ape (version 1.8-5)

NPRS.criterion: Objective Function Employed in Nonparametric Rate Smoothing

Description

NPRS.criterion computes the objective function to be minimized in the NPRS (nonparametric rate smoothing) algorithm described in Sanderson (1997).

Usage

NPRS.criterion(phy, chrono, expo = 2, minEdgeLength = 1e-06)

Arguments

phy

A non-clock-like phylogenetic tree (i.e. an object of class "phylo"), where the branch lengths are measured in substitutions.

chrono

A chronogram, i.e. a clock-like tree (i.e. an object of class "phylo"), where the branch lengths are measured in absolute time.

expo

Exponent in the objective function (default value: 2)

minEdgeLength

Minimum edge length in the phylogram (default value: 1e-06). If any branch lengths are smaller then they will be set to this value.

Value

• NPRS.criterion returns the value of the objective function given a phylogram and a chronogram.

Details

Please refer to Sanderson (1997) for mathematical details. Note that is is not computationally efficient to optimize the branch lengths in a chronogram by using NPRS.criterion - please use chronogram instead.

References

Sanderson, M. J. (1997) A nonparametric approach to estimating divergence times in the absence of rate constancy. Molecular Biology and Evolution, 14, 1218--1231.

See Also

ratogram, chronogram

Examples

library(ape)

# get tree
data("landplants.newick") # example tree in NH format
tree.landplants <- read.tree(text = landplants.newick)

# plot tree
tree.landplants
plot(tree.landplants, label.offset = 0.001)

# estimate chronogram
chrono.plants <- chronogram(tree.landplants)

# plot and write to file
plot(chrono.plants, label.offset = 0.001)
write.tree(chrono.plants, file = "chronogram.phy")

# value of NPRS function for our estimated chronogram
NPRS.criterion(tree.landplants, chrono.plants)